|Publication number||US7644408 B2|
|Application number||US 10/423,066|
|Publication date||Jan 5, 2010|
|Priority date||Apr 25, 2003|
|Also published as||US20040215590|
|Publication number||10423066, 423066, US 7644408 B2, US 7644408B2, US-B2-7644408, US7644408 B2, US7644408B2|
|Inventors||James L. Kroening|
|Original Assignee||Spotware Technologies, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (22), Non-Patent Citations (8), Referenced by (15), Classifications (11), Legal Events (2)|
|External Links: USPTO, USPTO Assignment, Espacenet|
1. Field of the Invention
The present invention relates to grid computing systems and more particularly pertains to a system for assigning and monitoring grid jobs on a computing grid.
2. Description of the Prior Art
Grid computing, which is sometimes referred to as distributed processing computing, has been proposed and explored as a means for bringing together a large number of computers of wide ranging locations and often disparate types for the purpose of utilizing idle computer processor time and/or unused storage by those needing processing or storage beyond their capabilities. While the development of public networks such as the Internet has facilitated communication between a wide range of computers all over the world, grid computing aims to facilitate not only communication between computers by also to coordinate processing by the computers in a useful manner. Typically, jobs are submitted to a managing entity of the grid system, and the job is executed by one or more of the grid computers making up the computing grid.
However, while the concept of grid computing holds great promise, the execution of the concept has not been without its challenges. One challenge associated with grid computing is matching grid jobs to the most appropriate grid computers on the computing grid and then tracking the progress of those grid jobs. Another challenge is tracking the many grid jobs that may simultaneously be running on various grid computers of the computing grid.
Grid jobs submitted to the computing grid may or may not have special requirements for the grid computers that perform the job. If there are special requirements, the grid manager must be informed of those requirements and then must sift through the profiles of the available grid computers to determine if that are any computers that meet the requirements, and if so, which of those computers meet the requirements. This can be time consuming, especially if there are a relatively large number of jobs that have special requirements, and thus can slow down the handling of the job before it even reaches a grid computer to perform the job.
Further, due to the desire to make the performance of the various grid operations on a grid computer as unobtrusive as possible to the primary user and primary functions of the grid computer, there may be a significant period of time before a grid job is begun, or an interruption that lasts for a significant period of time, and thus predicting the time when the grid job will be completed is difficult. The grid manager or grid job scheduler may query the grid computer as to the status of the grid job, but this may make grid operations on the grid computer more obtrusive and increase the traffic on the linking network. Further, if the network connection of the grid computer is intermittent, the grid manager may not be able to adequately time the queries to the periods when the network connection is present, and this may lead to a lack of response by the grid computer that confuses the grid manager as to the job status.
In view of the foregoing, it is believed that there is a significant need for a system that effectively matches the needs of the grid job requests to the abilities of the grid computers, and then effectively monitors the progress of the performance of the grid job.
In view of the foregoing, the present invention discloses a system for assigning and monitoring grid jobs on a computing grid.
One aspect of the invention involves a method and system for assigning a grid job to grid computers on a computing grid for performing the grid job. The method implementation comprises collecting information from the grid computers about attributes of the grid computer, grouping the grid computers of the computing grid into groups of grid computers based upon the attributes of the grid computers, receiving a grid job request from a grid customer, determining the attributes needed to perform the grid job request, comparing the attributes needed to perform the grid job to the attributes of the groups of the grid computers, and submitting the grid job to at least one of the grid computers of the group of grid computers having the attributes needed to perform the grid job. The system includes means for performing the steps of the method implementation.
Another aspect of the invention involves a method and system of reporting progress of performance of a grid job by a grid computer of a computing grid. The method implementation includes receiving a grid job request by the grid computer from a grid manager, initiating performance of the grid job request on the grid computer, periodically communicating a current status of the performance of the grid job request on the grid computer to the grid manager, and returning results of the performance of the grid job request to the grid manager. The system includes means for performing the steps of the method implementation.
Advantages of the invention, along with the various features of novelty which characterize the invention, are pointed out with particularity in the claims annexed to and forming a part of this disclosure. For a better understanding of the invention, its operating advantages and the specific objects attained by its uses, reference should be made to the accompanying drawings and descriptive matter in which there are illustrated preferred implementations of the invention.
The invention will be better understood and objects of the invention will become apparent when consideration is given to the following detailed description thereof. Such description makes reference to the annexed drawings wherein:
With reference now to the drawings, and in particular to
Initially, for the purposes of clarity in this description, terminology used throughout this description will be defined so as to minimize any confusion with respect to the disclosure of the invention, with the understanding that various other names may be given to the disclosed elements, and this terminology is not intended to be construed as limiting the invention.
A grid system 10 (see
In one embodiment of the invention, at least one of the grid computers 12 is located physically or geographically remote from at least one of the other grid computers, and in another embodiment, many or most of the grid computers are located physically or geographically remote from each other. The grid computers 12 and the grid manager computer 16 are linked in a manner suitable for permitting communication therebetween. The communication link between the computers may be a dedicated network, but also may be a public linking network such as the Internet.
In one aspect of the invention, a method is provided for steering or directing or assigning a grid job to the grid computers on a computing grid with capabilities that are highly suitable for performing the grid job (
Significantly, the grid computers of the computing grid may be grouped into a number of groups that each includes a plurality of grid computers (block 106). Each of the grid computers may be logically associated with one or more of the groups based upon the sharing of similar types or levels of one or more of the attributes reported to the grid manager during the registration process. This commonality may include having the same or similar levels of one or more of the various attributes recorded in the database so that the grid scheduler knows that each of the grid computers belonging to a particular group share the same attribute that is associated with the grouping, without having to check the profiles of each of the grid computers individually. For example, one grouping could include grid computers meeting a minimum level of security, while another grouping could include grid computers employing a particular operating system or specialized software application. Other groupings may include a relatively higher level of processor performance or availability to the grid, so that jobs submitted to computers of those groups will most likely be performed relatively quicker than jobs submitted to groups of computers with slower processors or less availability. The groupings are created logically by the grid manager, and are not the result of any physical connection or grouping of the computers, which may be located in locations highly remote to each other. A logical topology of each of the groupings of grid computers may be stored on the database (block 108) for future access by the grid manager or the grid scheduler. The topologies of the groupings may not be static, and may be updated as computers are added or removed from the computing grid, or as the configuration of the grid computers is altered.
As illustratively shown in
Significantly, the topologies of the groupings of the grid computers recorded on the database provide a guide as to which grid computers on the computing grid are most suitable for performing the grid job based upon the attributes shared by the grid computers of the particular groupings. The grid job scheduler may then avoid having to needlessly spend time comparing the attributes needed for performing the grid job to the attributes of each and every grid computer on the computing grid to determine suitable targets for the grid jobs. Further, subsequent similar grid job requests from the same grid customer may optionally be directed to the same grouping of computers without having to reanalyze the later grid job requests.
In another aspect of the invention, a method of reporting the progress of the performance of a grid job by one or more grid computers of the computing grid is provided (see
Significant to this aspect of the invention, the local manager application 20 keeps the grid manager informed of the progress of the performance of the grid job by communicating with the grid scheduler 22 periodically to report a current status of the grid job (block 144). The communication may occur even if the performance of the grid job has not begun, or even if the performance has not been completed. The communication between the local manager 20 and the grid job scheduler 22 may include sending a status message to the grid manager with the current status of the performance of the grid job. The status message may include information such as an amount or percentage of the job that has been completed or that remains incomplete (block 146). Preferably, the status message is relatively short and compact in size to minimize the imposition of on the grid computer and also on the grid manager, which may be handling such periodic messages from a multitude of computers on the computing grid. A short message also minimizes the traffic imposed upon the linking network. It is also desirable that the sending of the messages to the grid job scheduler is not dependent upon receiving inquiries from the grid job scheduler, or other computing grid entity, so that the local manager volunteers this information to the grid scheduler. This primarily one-way mode of communication lessens the quantity of message traffic between the grid computer and the grid manager, and also permits the grid scheduler to concentrate its inquiries on only the grid job assignments where status messages are the tardiest.
The communication of the job status messages may occur periodically during the performance of the grid job by the grid computer, and may even occur during periods of inactivity due to use of the grid computer by primary user to inform the grid scheduler that the grid job is alive and awaiting processor time to complete the job. The intervals between transmissions of the periodic messages may also be varied, such as by programming of the local manager by the grid scheduler, so that the grid job scheduler may be kept adequately updated on the progress of outstanding grid job requests until the grid job results are returned to the grid manager (block 148), but also so that the status messages are not too frequently sent to the scheduler. Further, if the grid scheduler has set the frequency for status messages, it is able to determine when the status messages are due, and when the messages are missed indicating that the performance of the grid job may have become hung up or terminated.
On the part of the grid manager (see
Upon completion of the grid job, the grid customer is informed when the grid job is complete and supplied with the results of the job (block 170).
The foregoing is considered as illustrative only of the principles of the invention. Further, since numerous modifications and changes will readily occur to those skilled in the art in view of the disclosure of this application, it is not desired to limit the invention to the exact embodiments, implementations, and operations shown and described. Accordingly, all equivalent relationships to those illustrated in the drawings and described in the specification, including all suitable modifications, are intended to be encompassed by the present invention that fall within the scope of the invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5881284 *||Oct 25, 1996||Mar 9, 1999||Nec Corporation||Method of scheduling a job in a clustered computer system and device therefor|
|US6070190||May 11, 1998||May 30, 2000||International Business Machines Corporation||Client-based application availability and response monitoring and reporting for distributed computing environments|
|US6243716||Aug 20, 1999||Jun 5, 2001||Sun Microsystems, Inc.||Methods and systems for distributed failure detection and recovery using leasing|
|US6338147||Oct 29, 1998||Jan 8, 2002||International Business Machines Corporation||Program products for performing checkpoint/restart of a parallel program|
|US6393583||Oct 29, 1998||May 21, 2002||International Business Machines Corporation||Method of performing checkpoint/restart of a parallel program|
|US6401216||Oct 29, 1998||Jun 4, 2002||International Business Machines Corporation||System of performing checkpoint/restart of a parallel program|
|US6415333||Dec 8, 1998||Jul 2, 2002||Telefonaktiebolaget L M Ericsson (Publ)||Distributed communication system with categorized resources|
|US6421704||Mar 20, 1998||Jul 16, 2002||Sun Microsystems, Inc.||Method, apparatus, and product for leasing of group membership in a distributed system|
|US6434594||Mar 9, 1999||Aug 13, 2002||Talk2 Technology, Inc.||Virtual processing network enabler|
|US6446126||May 12, 1999||Sep 3, 2002||Honeywell International Inc.||Ripple scheduling for end-to-end global resource management|
|US6597956 *||Aug 2, 2000||Jul 22, 2003||Terraspring, Inc.||Method and apparatus for controlling an extensible computing system|
|US7010596 *||Jun 28, 2002||Mar 7, 2006||International Business Machines Corporation||System and method for the allocation of grid computing to network workstations|
|US7065764 *||Jul 20, 2001||Jun 20, 2006||Netrendered, Inc.||Dynamically allocated cluster system|
|US7093004 *||Nov 27, 2002||Aug 15, 2006||Datasynapse, Inc.||Using execution statistics to select tasks for redundant assignment in a distributed computing platform|
|US7185046 *||Aug 28, 2002||Feb 27, 2007||Sun Microsystems, Inc.||Submitting jobs in a distributed computing environment|
|US20040019624 *||Jan 14, 2003||Jan 29, 2004||Hitachi, Ltd.||Computing system and control method|
|US20040044718 *||Aug 28, 2002||Mar 4, 2004||Ferstl Friedrich F.X.||Submitting jobs in a distributed computing environment|
|US20040093381 *||May 28, 2003||May 13, 2004||Hodges Donna Kay||Service-oriented architecture systems and methods|
|US20040098606 *||Nov 18, 2002||May 20, 2004||International Business Machines Corporation||System, method and program product for operating a grid of service providers based on a service policy|
|US20040225711 *||May 8, 2003||Nov 11, 2004||Spotware Technologies, Inc.||System for administering computers on a computing grid|
|US20050071843 *||Jan 24, 2002||Mar 31, 2005||Hong Guo||Topology aware scheduling for a multiprocessor system|
|US20060294238 *||Dec 16, 2002||Dec 28, 2006||Naik Vijay K||Policy-based hierarchical management of shared resources in a grid environment|
|1||*||Buyya et al., "Architectural Models for Resource Management in the Grid", Monash University, 2001.|
|2||DataGrid-WP01-Report on Current Technology Anglano, Ferrari, Giacomini, et al. Mar. 21, 2001.|
|3||Fundamentals of Grid Computing Redbooks Paper, IBM Corp Berstos. 2-2.|
|4||*||Kubicek et al., "Dynamic Allocation of Servers to Jobs in a Grid Hosting Enviroment", BT Technology Journal, vol. 22, No. 3, 2004, pp. 251-260.|
|5||*||Maheswaran et al., "Metagrid: A Scalable Famework for Wide-Area Service Deployment and Management", IEEE, 2002, pp. 1-8.|
|6||SunCluster Grid Architeture May 2002.|
|7||The Globus Heartbeat Monitor Specification v1.0 fp.globus.org/hbm/hearteat-spec.html.|
|8||The Globus Project: A Status Report Foster and Kesselman, 1998.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8166138 *||Jun 29, 2007||Apr 24, 2012||Apple Inc.||Network evaluation grid techniques|
|US8225319 *||Jun 27, 2006||Jul 17, 2012||Trimble Mrm Ltd.||Scheduling allocation of a combination of resources to a task that has a constraint|
|US8413153 *||Jun 12, 2009||Apr 2, 2013||Freescale Semiconductor Inc.||Methods and systems for sharing common job information|
|US8418183 *||Apr 9, 2013||Nec Corporation||Resource information collection and delivery method and system|
|US8484321||Mar 30, 2012||Jul 9, 2013||Apple Inc.||Network evaluation grid techniques|
|US9104478||Jun 15, 2012||Aug 11, 2015||Freescale Semiconductor, Inc.||System and method for improved job processing of a number of jobs belonging to communication streams within a data processor|
|US9286118||Jun 15, 2012||Mar 15, 2016||Freescale Semiconductor, Inc.||System and method for improved job processing to reduce contention for shared resources|
|US20060026556 *||Jul 29, 2005||Feb 2, 2006||Nec Corporation||Resource information collection and delivery method and system|
|US20060031509 *||Aug 5, 2005||Feb 9, 2006||Marco Ballette||Resource management method|
|US20070021998 *||Jun 27, 2006||Jan 25, 2007||Road Ltd.||Resource scheduling method and system|
|US20070271560 *||May 18, 2006||Nov 22, 2007||Microsoft Corporation||Deploying virtual machine to host based on workload characterizations|
|US20080270589 *||Mar 28, 2008||Oct 30, 2008||Begrid, Inc.||Multi-Source, Multi-Use Web Processing Through Dynamic Proxy Based Grid Computing Mechanisms|
|US20090006592 *||Jun 29, 2007||Jan 1, 2009||Novell, Inc.||Network evaluation grid techniques|
|US20100318996 *||Dec 16, 2010||Freescale Semiconductor, Inc.||Methods and systems for sharing common job information|
|US20130346993 *||Jun 20, 2012||Dec 26, 2013||Platform Computing Corporation||Job distribution within a grid environment|
|U.S. Classification||718/102, 709/201, 718/104, 709/226|
|International Classification||G06F9/50, G06F9/46, G06F15/16, G06F7/00, G06F15/173|
|Apr 25, 2003||AS||Assignment|
Owner name: SPOTWARE TECHNOLOGIES, INC., CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:KROENING, JAMES L.;REEL/FRAME:014010/0726
Effective date: 20030422
|Mar 11, 2013||FPAY||Fee payment|
Year of fee payment: 4